Pull f2fs updates from Jaegeuk Kim: "In this round, we've followed up to support some generic features such as cgroup, block reservation, linking fscrypt_ops, delivering write_hints, and some ioctls. And, we could fix some corner cases in terms of power-cut recovery and subtle deadlocks. Enhancements: - bitmap operations to handle NAT blocks - readahead to improve readdir speed - switch to use fscrypt_* - apply write hints for direct IO - add reserve_root=%u,resuid=%u,resgid=%u to reserve blocks for root/uid/gid - modify b_avail and b_free to consider root reserved blocks - support cgroup writeback - support FIEMAP_FLAG_XATTR for fibmap - add F2FS_IOC_PRECACHE_EXTENTS to pre-cache extents - add F2FS_IOC_{GET/SET}_PIN_FILE to pin LBAs for data blocks - support inode creation time Bug fixs: - sysfile-based quota operations - memory footprint accounting - allow to write data on partial preallocation case - fix deadlock case on fallocate - fix to handle fill_super errors - fix missing inode updates of fsync'ed file - recover renamed file which was fsycn'ed before - drop inmemory pages in corner error case - keep last_disk_size correctly - recover missing i_inline flags during roll-forward Various clean-up patches were added as well" Cherry-pick from origin/upstream-f2fs-stable-linux-4.4.y:5f9b3abb91
f2fs: support inode creation time9fb0de1751
f2fs: rebuild sit page from sit info in mem1062a0c018
f2fs: stop issuing discard if fs is readonlyfa043fae90
f2fs: clean up duplicated assignment in init_discard_policyb007190234
f2fs: use GFP_F2FS_ZERO for cleanup35b11839a1
f2fs: allow to recover node blocks given updated checkpointe56500860b
f2fs: recover some i_inline flags64aa9569a1
f2fs: correct removexattr behavior for null valued extended attribute70b3a923da
f2fs: drop page cache after fs shutdown8069a0e983
f2fs: stop gc/discard thread after fs shutdownbb924f7777
f2fs: hanlde error case in f2fs_ioc_shutdown700b53f21e
f2fs: split need_inplace_updatef31d52811c
f2fs: fix to update last_disk_size correctlyeeb0118b83
f2fs: kill F2FS_INLINE_XATTR_ADDRS for cleanupc1b74c9670
f2fs: clean up error path of fill_superd5efd57e01
f2fs: avoid hungtask when GC encrypted block if io_bits is setc4027d0843
f2fs: allow quota to use reserved blocks18d267c273
f2fs: fix to drop all inmem pages correctly4dca47531e
f2fs: speed up defragment on sparse file999f806a7c
f2fs: support F2FS_IOC_PRECACHE_EXTENTS84960fca96
f2fs: add an ioctl to disable GC for specific file292c8e1cfd
f2fs: prevent newly created inode from being dirtied incorrectly58b1f5b0fc
f2fs: support FIEMAP_FLAG_XATTR6afa9a94d0
f2fs: fix to cover f2fs_inline_data_fiemap with inode_lock10f4a4140b
f2fs: check node page again in write end iob203c58dfd
f2fs: fix to caclulate required free section correctlyd49132d45c
f2fs: handle newly created page when revoking inmem pages2ce6b9d816
f2fs: add resgid and resuid to reserve root blocksf53dcf6799
f2fs: implement cgroup writeback support1338f376d5
f2fs: remove unused pend_list_tagd4f19f6266
f2fs: avoid high cpu usage in discard threadb78e9302e2
f2fs: make local functions static62438ba87b
f2fs: add reserved blocks for root user06a366757f
f2fs: check segment type in __f2fs_replace_block4c6bc4be37
f2fs: update inode info to inode page for new file591b336387
f2fs: show precise # of blocks that user/root can useb242d7edc5
f2fs: clean up unneeded declaration87b8168e9e
f2fs: continue to do direct IO if we only preallocate partial blocks2b4d859bd9
f2fs: enable quota at remount from r to w54bf13a0ad
f2fs: skip stop_checkpoint for user data writes25ef3006ba
f2fs: fix missing error number for xattr operationcff2c7fe41
f2fs: recover directory operations by fsynce2bb618a0a
f2fs: return error during fill_super8a2c11d865
f2fs: fix an error case of missing update inode pagecd38d5ada5
f2fs: fix potential hangtask in f2fs_trace_pide81cafbeba
f2fs: no need return value in restore summary process04d44000d6
f2fs: use unlikely for release case925d0933d8
f2fs: don't return value in truncate_data_blocks_rangef7986c416d
f2fs: clean up f2fs_map_blockse4f5e26cda
f2fs: clean up hash codes1f994d4708
f2fs: fix error handling in fill_supere7db649b5f
f2fs: spread f2fs_k{m,z}alloc5d4e487b99
f2fs: inject fault to kvmalloc8b33886c37
f2fs: inject fault to kzallocd946807987
f2fs: remove a redundant conditional expression3bc01114a3
f2fs: apply write hints to select the type of segment for direct writec80f019591
f2fs: switch to fscrypt_prepare_setattr()bb8b850365
f2fs: switch to fscrypt_prepare_lookup()9ab470eaf8
f2fs: switch to fscrypt_prepare_rename()aeaac517a1
f2fs: switch to fscrypt_prepare_link()101c6a96ad
f2fs: switch to fscrypt_file_open()6d025237a1
f2fs: remove repeated f2fs_bug_onb01e03d724
f2fs: remove an excess variablee1f9be2f7c
f2fs: fix lock dependency in between dio_rwsem & i_mmap_seme5c7c86010
f2fs: remove unused parameterf130dbb98a
f2fs: still write data if preallocate only partial blocks47ee9b2598
f2fs: introduce sysfs readdir_ra to readahead inode block in readdir55e2f89181
f2fs: fix concurrent problem for updating free bitmape1398f6554
f2fs: remove unneeded memory footprint accounting2d69561135
f2fs: no need to read nat block if nat_block_bitmap is set4dd2d07338
f2fs: reserve nid resource for quota sysfile Signed-off-by: Jaegeuk Kim <jaegeuk@google.com>
162 lines
3.5 KiB
C
162 lines
3.5 KiB
C
/*
|
|
* f2fs IO tracer
|
|
*
|
|
* Copyright (c) 2014 Motorola Mobility
|
|
* Copyright (c) 2014 Jaegeuk Kim <jaegeuk@kernel.org>
|
|
*
|
|
* This program is free software; you can redistribute it and/or modify
|
|
* it under the terms of the GNU General Public License version 2 as
|
|
* published by the Free Software Foundation.
|
|
*/
|
|
#include <linux/fs.h>
|
|
#include <linux/f2fs_fs.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/radix-tree.h>
|
|
|
|
#include "f2fs.h"
|
|
#include "trace.h"
|
|
|
|
static RADIX_TREE(pids, GFP_ATOMIC);
|
|
static struct mutex pids_lock;
|
|
static struct last_io_info last_io;
|
|
|
|
static inline void __print_last_io(void)
|
|
{
|
|
if (!last_io.len)
|
|
return;
|
|
|
|
trace_printk("%3x:%3x %4x %-16s %2x %5x %5x %12x %4x\n",
|
|
last_io.major, last_io.minor,
|
|
last_io.pid, "----------------",
|
|
last_io.type,
|
|
last_io.fio.op, last_io.fio.op_flags,
|
|
last_io.fio.new_blkaddr,
|
|
last_io.len);
|
|
memset(&last_io, 0, sizeof(last_io));
|
|
}
|
|
|
|
static int __file_type(struct inode *inode, pid_t pid)
|
|
{
|
|
if (f2fs_is_atomic_file(inode))
|
|
return __ATOMIC_FILE;
|
|
else if (f2fs_is_volatile_file(inode))
|
|
return __VOLATILE_FILE;
|
|
else if (S_ISDIR(inode->i_mode))
|
|
return __DIR_FILE;
|
|
else if (inode->i_ino == F2FS_NODE_INO(F2FS_I_SB(inode)))
|
|
return __NODE_FILE;
|
|
else if (inode->i_ino == F2FS_META_INO(F2FS_I_SB(inode)))
|
|
return __META_FILE;
|
|
else if (pid)
|
|
return __NORMAL_FILE;
|
|
else
|
|
return __MISC_FILE;
|
|
}
|
|
|
|
void f2fs_trace_pid(struct page *page)
|
|
{
|
|
struct inode *inode = page->mapping->host;
|
|
pid_t pid = task_pid_nr(current);
|
|
void *p;
|
|
|
|
set_page_private(page, (unsigned long)pid);
|
|
|
|
if (radix_tree_preload(GFP_NOFS))
|
|
return;
|
|
|
|
mutex_lock(&pids_lock);
|
|
p = radix_tree_lookup(&pids, pid);
|
|
if (p == current)
|
|
goto out;
|
|
if (p)
|
|
radix_tree_delete(&pids, pid);
|
|
|
|
f2fs_radix_tree_insert(&pids, pid, current);
|
|
|
|
trace_printk("%3x:%3x %4x %-16s\n",
|
|
MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev),
|
|
pid, current->comm);
|
|
out:
|
|
mutex_unlock(&pids_lock);
|
|
radix_tree_preload_end();
|
|
}
|
|
|
|
void f2fs_trace_ios(struct f2fs_io_info *fio, int flush)
|
|
{
|
|
struct inode *inode;
|
|
pid_t pid;
|
|
int major, minor;
|
|
|
|
if (flush) {
|
|
__print_last_io();
|
|
return;
|
|
}
|
|
|
|
inode = fio->page->mapping->host;
|
|
pid = page_private(fio->page);
|
|
|
|
major = MAJOR(inode->i_sb->s_dev);
|
|
minor = MINOR(inode->i_sb->s_dev);
|
|
|
|
if (last_io.major == major && last_io.minor == minor &&
|
|
last_io.pid == pid &&
|
|
last_io.type == __file_type(inode, pid) &&
|
|
last_io.fio.op == fio->op &&
|
|
last_io.fio.op_flags == fio->op_flags &&
|
|
last_io.fio.new_blkaddr + last_io.len ==
|
|
fio->new_blkaddr) {
|
|
last_io.len++;
|
|
return;
|
|
}
|
|
|
|
__print_last_io();
|
|
|
|
last_io.major = major;
|
|
last_io.minor = minor;
|
|
last_io.pid = pid;
|
|
last_io.type = __file_type(inode, pid);
|
|
last_io.fio = *fio;
|
|
last_io.len = 1;
|
|
return;
|
|
}
|
|
|
|
void f2fs_build_trace_ios(void)
|
|
{
|
|
mutex_init(&pids_lock);
|
|
}
|
|
|
|
#define PIDVEC_SIZE 128
|
|
static unsigned int gang_lookup_pids(pid_t *results, unsigned long first_index,
|
|
unsigned int max_items)
|
|
{
|
|
struct radix_tree_iter iter;
|
|
void **slot;
|
|
unsigned int ret = 0;
|
|
|
|
if (unlikely(!max_items))
|
|
return 0;
|
|
|
|
radix_tree_for_each_slot(slot, &pids, &iter, first_index) {
|
|
results[ret] = iter.index;
|
|
if (++ret == max_items)
|
|
break;
|
|
}
|
|
return ret;
|
|
}
|
|
|
|
void f2fs_destroy_trace_ios(void)
|
|
{
|
|
pid_t pid[PIDVEC_SIZE];
|
|
pid_t next_pid = 0;
|
|
unsigned int found;
|
|
|
|
mutex_lock(&pids_lock);
|
|
while ((found = gang_lookup_pids(pid, next_pid, PIDVEC_SIZE))) {
|
|
unsigned idx;
|
|
|
|
next_pid = pid[found - 1] + 1;
|
|
for (idx = 0; idx < found; idx++)
|
|
radix_tree_delete(&pids, pid[idx]);
|
|
}
|
|
mutex_unlock(&pids_lock);
|
|
}
|